thumbnail

Performance evaluation of spectral vegetation indices using a statistical sensitivity function

Remote Sensing of Environment

By:
,
DOI: 10.1016/j.rse.2006.07.010

Links

Abstract

A great number of spectral vegetation indices (VIs) have been developed to estimate biophysical parameters of vegetation. Traditional techniques for evaluating the performance of VIs are regression-based statistics, such as the coefficient of determination and root mean square error. These statistics, however, are not capable of quantifying the detailed relationship between VIs and biophysical parameters because the sensitivity of a VI is usually a function of the biophysical parameter instead of a constant. To better quantify this relationship, we developed a "sensitivity function" for measuring the sensitivity of a VI to biophysical parameters. The sensitivity function is defined as the first derivative of the regression function, divided by the standard error of the dependent variable prediction. The function elucidates the change in sensitivity over the range of the biophysical parameter. The Student's t- or z-statistic can be used to test the significance of VI sensitivity. Additionally, we developed a "relative sensitivity function" that compares the sensitivities of two VIs when the biophysical parameters are unavailable. ?? 2006 Elsevier Inc. All rights reserved.

Additional Publication Details

Publication type:
Article
Publication Subtype:
Journal Article
Title:
Performance evaluation of spectral vegetation indices using a statistical sensitivity function
Series title:
Remote Sensing of Environment
DOI:
10.1016/j.rse.2006.07.010
Volume
106
Issue:
1
Year Published:
2007
Language:
English
Larger Work Type:
Article
Larger Work Subtype:
Journal Article
Larger Work Title:
Remote Sensing of Environment
First page:
59
Last page:
65
Number of Pages:
7